source("dados_playoffs.R")
## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.4 ✔ readr 2.1.5
## ✔ forcats 1.0.0 ✔ stringr 1.5.1
## ✔ lubridate 1.9.3 ✔ tibble 3.2.1
## ✔ purrr 1.0.2 ✔ tidyr 1.3.1
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
## Loading required package: splines
##
## Loading required package: gamlss.data
##
##
## Attaching package: 'gamlss.data'
##
##
## The following object is masked from 'package:datasets':
##
## sleep
##
##
## Loading required package: gamlss.dist
##
## Loading required package: nlme
##
##
## Attaching package: 'nlme'
##
##
## The following object is masked from 'package:dplyr':
##
## collapse
##
##
## Loading required package: parallel
##
## ********** GAMLSS Version 5.4-22 **********
##
## For more on GAMLSS look at https://www.gamlss.com/
##
## Type gamlssNews() to see new features/changes/bug fixes.
##
##
## Loading required package: carData
##
##
## Attaching package: 'car'
##
##
## The following object is masked from 'package:dplyr':
##
## recode
##
##
## The following object is masked from 'package:purrr':
##
## some
##
##
## Loading required package: zoo
##
##
## Attaching package: 'zoo'
##
##
## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numeric
######### Regressão linear ########
modelo_forwp <- lm(formula = WINP ~ PlusMinus + DREB, data = dados_regressaop)
modelo_forwp
##
## Call:
## lm(formula = WINP ~ PlusMinus + DREB, data = dados_regressaop)
##
## Coefficients:
## (Intercept) PlusMinus DREB
## 0.345213 0.025682 0.004033
coef(modelo_forwp)
## (Intercept) PlusMinus DREB
## 0.345212858 0.025681939 0.004033267
anova(modelo_forwp)
## Analysis of Variance Table
##
## Response: WINP
## Df Sum Sq Mean Sq F value Pr(>F)
## PlusMinus 1 7.3972 7.3972 681.7284 < 2e-16 ***
## DREB 1 0.0299 0.0299 2.7587 0.09805 .
## Residuals 237 2.5716 0.0109
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
summary(modelo_forwp) #Adjusted R-squared: 0.7406
##
## Call:
## lm(formula = WINP ~ PlusMinus + DREB, data = dados_regressaop)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.36469 -0.05721 0.01300 0.07160 0.36417
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.345213 0.078459 4.400 1.64e-05 ***
## PlusMinus 0.025682 0.001049 24.490 < 2e-16 ***
## DREB 0.004033 0.002428 1.661 0.098 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.1042 on 237 degrees of freedom
## Multiple R-squared: 0.7428, Adjusted R-squared: 0.7406
## F-statistic: 342.2 on 2 and 237 DF, p-value: < 2.2e-16
AIC(modelo_forwp)
## [1] -399.576
#### Resíduos Forward ###
plot(modelo_forwp, which = 1)
plot(modelo_forwp, which = 2)
plot(modelo_forwp, which = 3)
plot(modelo_forwp, which = 4)
plot(modelo_forwp, which = 5)
plot(modelo_forwp, which = 6)
shapiro.test(modelo_forwp$residuals) #p-value = 0.1997, normal
##
## Shapiro-Wilk normality test
##
## data: modelo_forwp$residuals
## W = 0.96952, p-value = 5.082e-05
#Teste de durbin watson para independencia
library(lmtest)
dwtest(modelo_forwp) #p-value = 0.07378
##
## Durbin-Watson test
##
## data: modelo_forwp
## DW = 1.8195, p-value = 0.07378
## alternative hypothesis: true autocorrelation is greater than 0
#Independência
plot(modelo_forwp$residuals,
ylab = "Residuos",
xlab = "Index dos Imovéis",
main = "Suposição de independência",
pch = 19)
#Homocedasticidade
plot(modelo_forwp$fitted.values, modelo_forwp$residuals,
xlab = "Valores Ajustados",
ylab = "Residuos",
pch = 19,
main = "Suposição de homocedasticidade"
)
#Breusch_Pagan para homocedasticdade
bptest(modelo_forwp) #p-value = 1.981e-05, heterocedasticidade
##
## studentized Breusch-Pagan test
##
## data: modelo_forwp
## BP = 21.659, df = 2, p-value = 1.981e-05
######## Betareg #######
### Logito #####
modelo_betapt_ftp <- betareg(formula = WINP_transformado ~ FTP + REB + PlusMinus, data = playoffs_transformado)
modelo_betapt_ftp
##
## Call:
## betareg(formula = WINP_transformado ~ FTP + REB + PlusMinus, data = playoffs_transformado)
##
## Coefficients (mean model with logit link):
## (Intercept) FTP REB PlusMinus
## -2.70112 0.01822 0.02605 0.14844
##
## Phi coefficients (precision model with identity link):
## (phi)
## 8.628
summary(modelo_betapt_ftp)
##
## Call:
## betareg(formula = WINP_transformado ~ FTP + REB + PlusMinus, data = playoffs_transformado)
##
## Standardized weighted residuals 2:
## Min 1Q Median 3Q Max
## -6.6072 -0.2417 0.2460 0.6519 1.5089
##
## Coefficients (mean model with logit link):
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.701118 0.960492 -2.812 0.00492 **
## FTP 0.018224 0.009515 1.915 0.05546 .
## REB 0.026050 0.013738 1.896 0.05793 .
## PlusMinus 0.148440 0.008141 18.233 < 2e-16 ***
##
## Phi coefficients (precision model with identity link):
## Estimate Std. Error z value Pr(>|z|)
## (phi) 8.6276 0.7677 11.24 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Type of estimator: ML (maximum likelihood)
## Log-likelihood: 168.6 on 5 Df
## Pseudo R-squared: 0.5202
## Number of iterations: 18 (BFGS) + 3 (Fisher scoring)
car::Anova(modelo_betapt_ftp)
## Analysis of Deviance Table (Type II tests)
##
## Response: WINP_transformado
## Df Chisq Pr(>Chisq)
## FTP 1 3.6683 0.05546 .
## REB 1 3.5957 0.05793 .
## PlusMinus 1 332.4251 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
coef(modelo_betapt_ftp)
## (Intercept) FTP REB PlusMinus (phi)
## -2.70111757 0.01822447 0.02605043 0.14844007 8.62758070
# Resíduos logito #
plot(modelo_betapt_ftp, which = 1)
plot(modelo_betapt_ftp, which = 2)
plot(modelo_betapt_ftp, which = 3)
plot(modelo_betapt_ftp, which = 4)
plot(modelo_betapt_ftp, which = 5) #QQplot não foi muito bom
plot(modelo_betapt_ftp, which = 6)
shapiro.test(modelo_betapt_ftp$residuals) #p-value = 0.7859, normal
##
## Shapiro-Wilk normality test
##
## data: modelo_betapt_ftp$residuals
## W = 0.99086, p-value = 0.1381
#Teste de durbin watson para independencia
library(lmtest)
dwtest(modelo_betapt_ftp) #p-value = 0.05838
##
## Durbin-Watson test
##
## data: modelo_betapt_ftp
## DW = 1.8033, p-value = 0.05838
## alternative hypothesis: true autocorrelation is greater than 0
#Independência
plot(modelo_betapt_ftp$residuals,
ylab = "Residuos",
xlab = "Index dos Imovéis",
main = "Suposição de independência",
pch = 19)
#Homocedasticidade
plot(modelo_betapt_ftp$fitted.values, modelo_betapt_ftp$residuals,
xlab = "Valores Ajustados",
ylab = "Residuos",
pch = 19,
main = "Suposição de homocedasticidade"
)
#Breusch_Pagan para homocedasticdade
bptest(modelo_betapt_ftp) #p-value = 0.0001505 heterocedasticidade
##
## studentized Breusch-Pagan test
##
## data: modelo_betapt_ftp
## BP = 20.252, df = 3, p-value = 0.0001505
### loglog ####
modelop_loglog_reb <- betareg(formula = WINP_transformado ~ REB + PlusMinus, data = playoffs_transformado,link = "loglog")
modelop_loglog_reb
##
## Call:
## betareg(formula = WINP_transformado ~ REB + PlusMinus, data = playoffs_transformado,
## link = "loglog")
##
## Coefficients (mean model with loglog link):
## (Intercept) REB PlusMinus
## -0.64302 0.02108 0.08192
##
## Phi coefficients (precision model with identity link):
## (phi)
## 8.439
summary(modelop_loglog_reb)
##
## Call:
## betareg(formula = WINP_transformado ~ REB + PlusMinus, data = playoffs_transformado,
## link = "loglog")
##
## Standardized weighted residuals 2:
## Min 1Q Median 3Q Max
## -6.8898 -0.1969 0.2649 0.6177 1.5457
##
## Coefficients (mean model with loglog link):
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -0.643022 0.325933 -1.973 0.04851 *
## REB 0.021083 0.007695 2.740 0.00615 **
## PlusMinus 0.081916 0.003444 23.784 < 2e-16 ***
##
## Phi coefficients (precision model with identity link):
## Estimate Std. Error z value Pr(>|z|)
## (phi) 8.4390 0.7512 11.23 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Type of estimator: ML (maximum likelihood)
## Log-likelihood: 162.3 on 4 Df
## Pseudo R-squared: 0.6689
## Number of iterations: 15 (BFGS) + 1 (Fisher scoring)
car::Anova(modelop_loglog_reb)
## Analysis of Deviance Table (Type II tests)
##
## Response: WINP_transformado
## Df Chisq Pr(>Chisq)
## REB 1 7.5065 0.006148 **
## PlusMinus 1 565.6606 < 2.2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
coef(modelop_loglog_reb)
## (Intercept) REB PlusMinus (phi)
## -0.64302232 0.02108259 0.08191571 8.43898300
# Resíduos logito #
plot(modelop_loglog_reb, which = 1)
plot(modelop_loglog_reb, which = 2)
plot(modelop_loglog_reb, which = 3)
plot(modelop_loglog_reb, which = 4)
plot(modelop_loglog_reb, which = 5) #QQplot não foi muito bom
plot(modelop_loglog_reb, which = 6)
shapiro.test(modelop_loglog_reb$residuals) #p-value = 0.005181, normal
##
## Shapiro-Wilk normality test
##
## data: modelop_loglog_reb$residuals
## W = 0.98276, p-value = 0.005181
#Teste de durbin watson para independencia
library(lmtest)
dwtest(modelop_loglog_reb) #p-value = 0.07596
##
## Durbin-Watson test
##
## data: modelop_loglog_reb
## DW = 1.8197, p-value = 0.07596
## alternative hypothesis: true autocorrelation is greater than 0
#Independência
plot(modelop_loglog_reb$residuals,
ylab = "Residuos",
xlab = "Index dos Imovéis",
main = "Suposição de independência",
pch = 19)
#Homocedasticidade
plot(modelop_loglog_reb$fitted.values, modelop_loglog_reb$residuals,
xlab = "Valores Ajustados",
ylab = "Residuos",
pch = 19,
main = "Suposição de homocedasticidade"
)
#Breusch_Pagan para homocedasticdade
bptest(modelop_loglog_reb) #p-value = 4.637e-05 heterocedasticidade
##
## studentized Breusch-Pagan test
##
## data: modelop_loglog_reb
## BP = 19.958, df = 2, p-value = 4.637e-05
### Probito #####
modelop_probit_ftp <- betareg(formula = WINP_transformado ~ FTP + REB + PlusMinus, data = playoffs_transformado,
link = "probit")
modelop_probit_ftp
##
## Call:
## betareg(formula = WINP_transformado ~ FTP + REB + PlusMinus, data = playoffs_transformado,
## link = "probit")
##
## Coefficients (mean model with probit link):
## (Intercept) FTP REB PlusMinus
## -1.61947 0.01013 0.01697 0.08691
##
## Phi coefficients (precision model with identity link):
## (phi)
## 8.695
summary(modelop_probit_ftp)
##
## Call:
## betareg(formula = WINP_transformado ~ FTP + REB + PlusMinus, data = playoffs_transformado,
## link = "probit")
##
## Standardized weighted residuals 2:
## Min 1Q Median 3Q Max
## -6.7399 -0.2532 0.2570 0.6393 1.5064
##
## Coefficients (mean model with probit link):
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.619472 0.557171 -2.907 0.00365 **
## FTP 0.010132 0.005527 1.833 0.06679 .
## REB 0.016972 0.008038 2.111 0.03475 *
## PlusMinus 0.086908 0.004332 20.061 < 2e-16 ***
##
## Phi coefficients (precision model with identity link):
## Estimate Std. Error z value Pr(>|z|)
## (phi) 8.6950 0.7759 11.21 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Type of estimator: ML (maximum likelihood)
## Log-likelihood: 168.2 on 5 Df
## Pseudo R-squared: 0.5897
## Number of iterations: 18 (BFGS) + 2 (Fisher scoring)
car::Anova(modelop_probit_ftp)
## Analysis of Deviance Table (Type II tests)
##
## Response: WINP_transformado
## Df Chisq Pr(>Chisq)
## FTP 1 3.3603 0.06679 .
## REB 1 4.4576 0.03475 *
## PlusMinus 1 402.4243 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
coef(modelop_probit_ftp)
## (Intercept) FTP REB PlusMinus (phi)
## -1.61947183 0.01013154 0.01697157 0.08690779 8.69502494
# Resíduos logito #
plot(modelop_probit_ftp, which = 1)
plot(modelop_probit_ftp, which = 2)
plot(modelop_probit_ftp, which = 3)
plot(modelop_probit_ftp, which = 4)
plot(modelop_probit_ftp, which = 5) #QQplot não foi muito bom
plot(modelop_probit_ftp, which = 6)
shapiro.test(modelop_probit_ftp$residuals) #p-value = 0.08389, normal
##
## Shapiro-Wilk normality test
##
## data: modelop_probit_ftp$residuals
## W = 0.98965, p-value = 0.08389
#Teste de durbin watson para independencia
library(lmtest)
dwtest(modelop_probit_ftp) #p-value = 0.05838
##
## Durbin-Watson test
##
## data: modelop_probit_ftp
## DW = 1.8033, p-value = 0.05838
## alternative hypothesis: true autocorrelation is greater than 0
#Independência
plot(modelop_probit_ftp$residuals,
ylab = "Residuos",
xlab = "Index dos Imovéis",
main = "Suposição de independência",
pch = 19)
#Homocedasticidade
plot(modelop_probit_ftp$fitted.values, modelop_probit_ftp$residuals,
xlab = "Valores Ajustados",
ylab = "Residuos",
pch = 19,
main = "Suposição de homocedasticidade"
)
#Breusch_Pagan para homocedasticdade
bptest(modelop_probit_ftp) #p-value = 0.0001505 heterocedasticidade
##
## studentized Breusch-Pagan test
##
## data: modelop_probit_ftp
## BP = 20.252, df = 3, p-value = 0.0001505
### cloglog ####
modelo_betat_cloglog_ftp <- betareg(formula = WINP_transformado ~ FTP + REB + PlusMinus, data = playoffs_transformado,
link = "cloglog")
modelo_betat_cloglog_ftp
##
## Call:
## betareg(formula = WINP_transformado ~ FTP + REB + PlusMinus, data = playoffs_transformado,
## link = "cloglog")
##
## Coefficients (mean model with cloglog link):
## (Intercept) FTP REB PlusMinus
## -2.51458 0.01496 0.01872 0.11192
##
## Phi coefficients (precision model with identity link):
## (phi)
## 8.037
summary(modelo_betat_cloglog_ftp)
##
## Call:
## betareg(formula = WINP_transformado ~ FTP + REB + PlusMinus, data = playoffs_transformado,
## link = "cloglog")
##
## Standardized weighted residuals 2:
## Min 1Q Median 3Q Max
## -6.2025 -0.2494 0.3180 0.7076 1.4128
##
## Coefficients (mean model with cloglog link):
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -2.514577 0.766704 -3.280 0.00104 **
## FTP 0.014962 0.007520 1.989 0.04665 *
## REB 0.018717 0.010768 1.738 0.08218 .
## PlusMinus 0.111920 0.006211 18.021 < 2e-16 ***
##
## Phi coefficients (precision model with identity link):
## Estimate Std. Error z value Pr(>|z|)
## (phi) 8.0370 0.7113 11.3 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Type of estimator: ML (maximum likelihood)
## Log-likelihood: 162.2 on 5 Df
## Pseudo R-squared: 0.4689
## Number of iterations: 19 (BFGS) + 1 (Fisher scoring)
car::Anova(modelo_betat_cloglog_ftp)
## Analysis of Deviance Table (Type II tests)
##
## Response: WINP_transformado
## Df Chisq Pr(>Chisq)
## FTP 1 3.9579 0.04665 *
## REB 1 3.0212 0.08218 .
## PlusMinus 1 324.7531 < 2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
coef(modelo_betat_cloglog_ftp)
## (Intercept) FTP REB PlusMinus (phi)
## -2.51457700 0.01496152 0.01871662 0.11191991 8.03696860
# Resíduos logito #
plot(modelo_betat_cloglog_ftp, which = 1)
plot(modelo_betat_cloglog_ftp, which = 2)
plot(modelo_betat_cloglog_ftp, which = 3)
plot(modelo_betat_cloglog_ftp, which = 4)
plot(modelo_betat_cloglog_ftp, which = 5) #QQplot não foi muito bom
plot(modelo_betat_cloglog_ftp, which = 6)
shapiro.test(modelo_betat_cloglog_ftp$residuals) #p-value = 0.08389, normal
##
## Shapiro-Wilk normality test
##
## data: modelo_betat_cloglog_ftp$residuals
## W = 0.987, p-value = 0.02824
#Teste de durbin watson para independencia
library(lmtest)
dwtest(modelo_betat_cloglog_ftp) #p-value = 0.05838
##
## Durbin-Watson test
##
## data: modelo_betat_cloglog_ftp
## DW = 1.8033, p-value = 0.05838
## alternative hypothesis: true autocorrelation is greater than 0
#Independência
plot(modelo_betat_cloglog_ftp$residuals,
ylab = "Residuos",
xlab = "Index dos Imovéis",
main = "Suposição de independência",
pch = 19)
#Homocedasticidade
plot(modelo_betat_cloglog_ftp$fitted.values, modelo_betat_cloglog_ftp$residuals,
xlab = "Valores Ajustados",
ylab = "Residuos",
pch = 19,
main = "Suposição de homocedasticidade"
)
#Breusch_Pagan para homocedasticdade
bptest(modelo_betat_cloglog_ftp) #p-value = 0.0001505 heterocedasticidade
##
## studentized Breusch-Pagan test
##
## data: modelo_betat_cloglog_ftp
## BP = 20.252, df = 3, p-value = 0.0001505
### cauchito ####
modelo_betat_cauchit_ftp <- betareg(formula = WINP_transformado ~ FTP + PlusMinus, data = playoffs_transformado,
link = "cauchit")
modelo_betat_cauchit_ftp
##
## Call:
## betareg(formula = WINP_transformado ~ FTP + PlusMinus, data = playoffs_transformado,
## link = "cauchit")
##
## Coefficients (mean model with cauchit link):
## (Intercept) FTP PlusMinus
## -1.58499 0.01838 0.15092
##
## Phi coefficients (precision model with identity link):
## (phi)
## 7.34
summary(modelo_betat_cauchit_ftp)
##
## Call:
## betareg(formula = WINP_transformado ~ FTP + PlusMinus, data = playoffs_transformado,
## link = "cauchit")
##
## Standardized weighted residuals 2:
## Min 1Q Median 3Q Max
## -5.7092 -0.3498 0.2333 0.6419 1.3118
##
## Coefficients (mean model with cauchit link):
## Estimate Std. Error z value Pr(>|z|)
## (Intercept) -1.58499 0.79614 -1.991 0.0465 *
## FTP 0.01838 0.01033 1.780 0.0751 .
## PlusMinus 0.15092 0.01172 12.882 <2e-16 ***
##
## Phi coefficients (precision model with identity link):
## Estimate Std. Error z value Pr(>|z|)
## (phi) 7.3404 0.6427 11.42 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Type of estimator: ML (maximum likelihood)
## Log-likelihood: 154.6 on 4 Df
## Pseudo R-squared: 0.2574
## Number of iterations: 60 (BFGS) + 2 (Fisher scoring)
car::Anova(modelo_betat_cauchit_ftp)
## Analysis of Deviance Table (Type II tests)
##
## Response: WINP_transformado
## Df Chisq Pr(>Chisq)
## FTP 1 3.1679 0.0751 .
## PlusMinus 1 165.9406 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
coef(modelo_betat_cauchit_ftp)
## (Intercept) FTP PlusMinus (phi)
## -1.58499016 0.01838294 0.15091652 7.34042352
# Resíduos logito #
plot(modelo_betat_cauchit_ftp, which = 1)
plot(modelo_betat_cauchit_ftp, which = 2)
plot(modelo_betat_cauchit_ftp, which = 3)
plot(modelo_betat_cauchit_ftp, which = 4)
plot(modelo_betat_cauchit_ftp, which = 5) #QQplot não foi muito bom
plot(modelo_betat_cauchit_ftp, which = 6)
shapiro.test(modelo_betat_cauchit_ftp$residuals) #p-value = 0.05594, normal
##
## Shapiro-Wilk normality test
##
## data: modelo_betat_cauchit_ftp$residuals
## W = 0.98867, p-value = 0.05594
#Teste de durbin watson para independencia
library(lmtest)
dwtest(modelo_betat_cauchit_ftp) #p-value = 0.06737
##
## Durbin-Watson test
##
## data: modelo_betat_cauchit_ftp
## DW = 1.8122, p-value = 0.06737
## alternative hypothesis: true autocorrelation is greater than 0
#Independência
plot(modelo_betat_cauchit_ftp$residuals,
ylab = "Residuos",
xlab = "Index dos Imovéis",
main = "Suposição de independência",
pch = 19)
#Homocedasticidade
plot(modelo_betat_cauchit_ftp$fitted.values, modelo_betat_cauchit_ftp$residuals,
xlab = "Valores Ajustados",
ylab = "Residuos",
pch = 19,
main = "Suposição de homocedasticidade"
)
#Breusch_Pagan para homocedasticdade
bptest(modelo_betat_cauchit_ftp) #p-value = 0.0001374 heterocedasticidade
##
## studentized Breusch-Pagan test
##
## data: modelo_betat_cauchit_ftp
## BP = 17.785, df = 2, p-value = 0.0001374
######## Gamlss #######
### Beta ####
gamlss_betap_pf <- gamlss(formula = WINP ~ PF + PlusMinus, family = BEZI, data = dados_regressaop)
## GAMLSS-RS iteration 1: Global Deviance = -161.4236
## GAMLSS-RS iteration 2: Global Deviance = -329.2529
## GAMLSS-RS iteration 3: Global Deviance = -330.669
## GAMLSS-RS iteration 4: Global Deviance = -330.6692
gamlss_betap_pf
##
## Family: c("BEZI", "Zero Inflated Beta")
## Fitting method: RS()
##
## Call: gamlss(formula = WINP ~ PF + PlusMinus, family = BEZI,
## data = dados_regressaop)
##
## Mu Coefficients:
## (Intercept) PF PlusMinus
## 0.5145 -0.0265 0.1063
## Sigma Coefficients:
## (Intercept)
## 3.571
## Nu Coefficients:
## (Intercept)
## -2.197
##
## Degrees of Freedom for the fit: 5 Residual Deg. of Freedom 235
## Global Deviance: -330.669
## AIC: -320.669
## SBC: -303.266
summary(gamlss_betap_pf)
## ******************************************************************
## Family: c("BEZI", "Zero Inflated Beta")
##
## Call: gamlss(formula = WINP ~ PF + PlusMinus, family = BEZI,
## data = dados_regressaop)
##
## Fitting method: RS()
##
## ------------------------------------------------------------------
## Mu link function: logit
## Mu Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.514484 0.244311 2.106 0.0363 *
## PF -0.026499 0.011248 -2.356 0.0193 *
## PlusMinus 0.106287 0.004552 23.349 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ------------------------------------------------------------------
## Sigma link function: log
## Sigma Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.57084 0.09504 37.57 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ------------------------------------------------------------------
## Nu link function: logit
## Nu Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -2.1972 0.2152 -10.21 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ------------------------------------------------------------------
## No. of observations in the fit: 240
## Degrees of Freedom for the fit: 5
## Residual Deg. of Freedom: 235
## at cycle: 4
##
## Global Deviance: -330.6692
## AIC: -320.6692
## SBC: -303.266
## ******************************************************************
coef(gamlss_betap_pf)
## (Intercept) PF PlusMinus
## 0.51448416 -0.02649885 0.10628690
# Resíduos logito #
plot(gamlss_betap_pf, which = 1)
## ******************************************************************
## Summary of the Randomised Quantile Residuals
## mean = -0.1698301
## variance = 0.7175177
## coef. of skewness = -0.5060405
## coef. of kurtosis = 2.606695
## Filliben correlation coefficient = 0.9838817
## ******************************************************************
shapiro.test(gamlss_betap_pf$residuals) #p-value = 0.05594, normal
##
## Shapiro-Wilk normality test
##
## data: gamlss_betap_pf$residuals
## W = 0.96679, p-value = 2.193e-05
#Teste de durbin watson para independencia
library(lmtest)
dwtest(gamlss_betap_pf) #p-value = 0.06737
##
## Durbin-Watson test
##
## data: gamlss_betap_pf
## DW = 1.8508, p-value = 0.1176
## alternative hypothesis: true autocorrelation is greater than 0
#Independência
plot(gamlss_betap_pf$residuals,
ylab = "Residuos",
xlab = "Index dos Imovéis",
main = "Suposição de independência",
pch = 19)
#Breusch_Pagan para homocedasticdade
bptest(gamlss_betap_pf) #p-value = 0.0001374 heterocedasticidade
##
## studentized Breusch-Pagan test
##
## data: gamlss_betap_pf
## BP = 18.652, df = 2, p-value = 8.91e-05
######## Modelos Mistos #######
##### Normal Team ####
misto_normalp_dreb <- gamlss(formula = WINP ~ (re(random = ~1 | TEAM)) +
PlusMinus + DREB, family = NO, data = dados_regressaop)
## GAMLSS-RS iteration 1: Global Deviance = -436.3104
## GAMLSS-RS iteration 2: Global Deviance = -436.3104
misto_normalp_dreb
##
## Family: c("NO", "Normal")
## Fitting method: RS()
##
## Call: gamlss(formula = WINP ~ (re(random = ~1 | TEAM)) +
## PlusMinus + DREB, family = NO, data = dados_regressaop)
##
## Mu Coefficients:
## (Intercept) re(random = ~1 | TEAM) PlusMinus
## 0.343406 NA 0.025386
## DREB
## 0.004065
## Sigma Coefficients:
## (Intercept)
## -2.328
##
## Degrees of Freedom for the fit: 14.43 Residual Deg. of Freedom 225.6
## Global Deviance: -436.31
## AIC: -407.444
## SBC: -357.207
coef(misto_normalp_dreb)
## (Intercept) re(random = ~1 | TEAM) PlusMinus
## 0.343405846 NA 0.025385907
## DREB
## 0.004064528
summary(misto_normalp_dreb) #AIC:
## ******************************************************************
## Family: c("NO", "Normal")
##
## Call: gamlss(formula = WINP ~ (re(random = ~1 | TEAM)) +
## PlusMinus + DREB, family = NO, data = dados_regressaop)
##
## Fitting method: RS()
##
## ------------------------------------------------------------------
## Mu link function: identity
## Mu Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.3434058 0.0734368 4.676 5.03e-06 ***
## PlusMinus 0.0253859 0.0009815 25.864 < 2e-16 ***
## DREB 0.0040645 0.0022729 1.788 0.0751 .
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ------------------------------------------------------------------
## Sigma link function: log
## Sigma Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -2.32792 0.04564 -51 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ------------------------------------------------------------------
## NOTE: Additive smoothing terms exist in the formulas:
## i) Std. Error for smoothers are for the linear effect only.
## ii) Std. Error for the linear terms maybe are not accurate.
## ------------------------------------------------------------------
## No. of observations in the fit: 240
## Degrees of Freedom for the fit: 14.43325
## Residual Deg. of Freedom: 225.5668
## at cycle: 2
##
## Global Deviance: -436.3104
## AIC: -407.4439
## SBC: -357.207
## ******************************************************************
getSmo(misto_normalp_dreb)
## Linear mixed-effects model fit by maximum likelihood
## Data: Data
## Log-likelihood: 204.9898
## Fixed: fix.formula
## (Intercept)
## -0.0008263643
##
## Random effects:
## Formula: ~1 | TEAM
## (Intercept) Residual
## StdDev: 0.0282907 1.024705
##
## Variance function:
## Structure: fixed weights
## Formula: ~W.var
## Number of Observations: 240
## Number of Groups: 33
#Resíduos
plot(misto_normalp_dreb)
## ******************************************************************
## Summary of the Quantile Residuals
## mean = -4.491375e-17
## variance = 1.004184
## coef. of skewness = -0.5315828
## coef. of kurtosis = 3.924077
## Filliben correlation coefficient = 0.9864969
## ******************************************************************
shapiro.test(misto_normalp_dreb$residuals) #p-value =
##
## Shapiro-Wilk normality test
##
## data: misto_normalp_dreb$residuals
## W = 0.97414, p-value = 0.000228
#Independência
plot(misto_normalp_dreb$residuals,
ylab = "Residuos",
xlab = "Index dos Imovéis",
main = "Suposição de independência",
pch = 19)
#Breusch_Pagan para homocedasticdade
bptest(misto_normalp_dreb) #p-value =
##
## studentized Breusch-Pagan test
##
## data: misto_normalp_dreb
## BP = 21.659, df = 2, p-value = 1.981e-05
##### Normal Temporada ####
misto_normalp_temp_team <- gamlss(formula = WINP ~ (re(random = ~1 | Numero_temporada)) +
PlusMinus + DREB + TEAM, family = NO, data = dados_regressaop)
## GAMLSS-RS iteration 1: Global Deviance = -473.5697
## GAMLSS-RS iteration 2: Global Deviance = -473.5697
misto_normalp_temp_team
##
## Family: c("NO", "Normal")
## Fitting method: RS()
##
## Call: gamlss(formula = WINP ~ (re(random = ~1 | Numero_temporada)) +
## PlusMinus + DREB + TEAM, family = NO, data = dados_regressaop)
##
## Mu Coefficients:
## (Intercept) re(random = ~1 | Numero_temporada)
## 0.346627 NA
## PlusMinus DREB
## 0.024781 0.004688
## TEAMBoston Celtics TEAMBrooklyn Nets
## -0.005755 -0.103945
## TEAMCharlotte Bobcats TEAMCharlotte Hornets
## -0.249427 0.163699
## TEAMChicago Bulls TEAMCleveland Cavaliers
## -0.035875 0.021331
## TEAMDallas Mavericks TEAMDenver Nuggets
## -0.021046 -0.046537
## TEAMDetroit Pistons TEAMGolden State Warriors
## -0.089758 0.034484
## TEAMHouston Rockets TEAMIndiana Pacers
## 0.011700 -0.130056
## TEAMLA Clippers TEAMLos Angeles Clippers
## -0.068940 -0.030764
## TEAMLos Angeles Lakers TEAMMemphis Grizzlies
## 0.012975 0.018290
## TEAMMiami Heat TEAMMilwaukee Bucks
## 0.019437 -0.036961
## TEAMMinnesota Timberwolves TEAMNew Orleans Hornets
## -0.068361 0.183304
## TEAMNew Orleans Pelicans TEAMNew York Knicks
## -0.120687 -0.067320
## TEAMOklahoma City Thunder TEAMOrlando Magic
## -0.039946 -0.038859
## TEAMPhiladelphia 76ers TEAMPhoenix Suns
## -0.029856 0.056248
## TEAMPortland Trail Blazers TEAMSacramento Kings
## -0.036629 -0.023561
## TEAMSan Antonio Spurs TEAMToronto Raptors
## -0.043036 0.015121
## TEAMUtah Jazz TEAMWashington Wizards
## -0.046544 -0.002517
## Sigma Coefficients:
## (Intercept)
## -2.406
##
## Degrees of Freedom for the fit: 35 Residual Deg. of Freedom 205
## Global Deviance: -473.57
## AIC: -403.57
## SBC: -281.747
coef(misto_normalp_temp_team)
## (Intercept) re(random = ~1 | Numero_temporada)
## 0.346626952 NA
## PlusMinus DREB
## 0.024780875 0.004687656
## TEAMBoston Celtics TEAMBrooklyn Nets
## -0.005755339 -0.103944840
## TEAMCharlotte Bobcats TEAMCharlotte Hornets
## -0.249427359 0.163699455
## TEAMChicago Bulls TEAMCleveland Cavaliers
## -0.035874677 0.021330971
## TEAMDallas Mavericks TEAMDenver Nuggets
## -0.021046376 -0.046537067
## TEAMDetroit Pistons TEAMGolden State Warriors
## -0.089758350 0.034483655
## TEAMHouston Rockets TEAMIndiana Pacers
## 0.011700011 -0.130055911
## TEAMLA Clippers TEAMLos Angeles Clippers
## -0.068939682 -0.030764280
## TEAMLos Angeles Lakers TEAMMemphis Grizzlies
## 0.012975042 0.018290493
## TEAMMiami Heat TEAMMilwaukee Bucks
## 0.019437176 -0.036960816
## TEAMMinnesota Timberwolves TEAMNew Orleans Hornets
## -0.068361098 0.183303607
## TEAMNew Orleans Pelicans TEAMNew York Knicks
## -0.120686534 -0.067319855
## TEAMOklahoma City Thunder TEAMOrlando Magic
## -0.039945586 -0.038859328
## TEAMPhiladelphia 76ers TEAMPhoenix Suns
## -0.029855977 0.056248385
## TEAMPortland Trail Blazers TEAMSacramento Kings
## -0.036628537 -0.023561117
## TEAMSan Antonio Spurs TEAMToronto Raptors
## -0.043035574 0.015121085
## TEAMUtah Jazz TEAMWashington Wizards
## -0.046544409 -0.002516683
summary(misto_normalp_temp_team) #AIC:
## ******************************************************************
## Family: c("NO", "Normal")
##
## Call: gamlss(formula = WINP ~ (re(random = ~1 | Numero_temporada)) +
## PlusMinus + DREB + TEAM, family = NO, data = dados_regressaop)
##
## Fitting method: RS()
##
## ------------------------------------------------------------------
## Mu link function: identity
## Mu Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.346627 0.076655 4.522 1.04e-05 ***
## PlusMinus 0.024781 0.001032 24.023 < 2e-16 ***
## DREB 0.004688 0.002300 2.038 0.042839 *
## TEAMBoston Celtics -0.005755 0.035805 -0.161 0.872455
## TEAMBrooklyn Nets -0.103945 0.041208 -2.522 0.012413 *
## TEAMCharlotte Bobcats -0.249427 0.069121 -3.609 0.000387 ***
## TEAMCharlotte Hornets 0.163699 0.094029 1.741 0.083193 .
## TEAMChicago Bulls -0.035875 0.039991 -0.897 0.370738
## TEAMCleveland Cavaliers 0.021331 0.043561 0.490 0.624881
## TEAMDallas Mavericks -0.021046 0.038632 -0.545 0.586486
## TEAMDenver Nuggets -0.046537 0.038789 -1.200 0.231619
## TEAMDetroit Pistons -0.089758 0.059407 -1.511 0.132354
## TEAMGolden State Warriors 0.034484 0.040998 0.841 0.401271
## TEAMHouston Rockets 0.011700 0.040066 0.292 0.770568
## TEAMIndiana Pacers -0.130056 0.039828 -3.265 0.001281 **
## TEAMLA Clippers -0.068940 0.045216 -1.525 0.128885
## TEAMLos Angeles Clippers -0.030764 0.052216 -0.589 0.556392
## TEAMLos Angeles Lakers 0.012975 0.041296 0.314 0.753693
## TEAMMemphis Grizzlies 0.018290 0.038643 0.473 0.636485
## TEAMMiami Heat 0.019437 0.036987 0.526 0.599798
## TEAMMilwaukee Bucks -0.036961 0.039210 -0.943 0.346979
## TEAMMinnesota Timberwolves -0.068361 0.058750 -1.164 0.245944
## TEAMNew Orleans Hornets 0.183304 0.070151 2.613 0.009641 **
## TEAMNew Orleans Pelicans -0.120687 0.058469 -2.064 0.040267 *
## TEAMNew York Knicks -0.067320 0.048063 -1.401 0.162832
## TEAMOklahoma City Thunder -0.039946 0.039161 -1.020 0.308916
## TEAMOrlando Magic -0.038859 0.045153 -0.861 0.390456
## TEAMPhiladelphia 76ers -0.029856 0.040094 -0.745 0.457336
## TEAMPhoenix Suns 0.056248 0.052393 1.074 0.284267
## TEAMPortland Trail Blazers -0.036629 0.037850 -0.968 0.334323
## TEAMSacramento Kings -0.023561 0.094120 -0.250 0.802582
## TEAMSan Antonio Spurs -0.043036 0.038012 -1.132 0.258893
## TEAMToronto Raptors 0.015121 0.041208 0.367 0.714036
## TEAMUtah Jazz -0.046544 0.040055 -1.162 0.246577
## TEAMWashington Wizards -0.002517 0.048267 -0.052 0.958467
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ------------------------------------------------------------------
## Sigma link function: log
## Sigma Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -2.40554 0.04564 -52.7 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ------------------------------------------------------------------
## NOTE: Additive smoothing terms exist in the formulas:
## i) Std. Error for smoothers are for the linear effect only.
## ii) Std. Error for the linear terms maybe are not accurate.
## ------------------------------------------------------------------
## No. of observations in the fit: 240
## Degrees of Freedom for the fit: 35
## Residual Deg. of Freedom: 205
## at cycle: 2
##
## Global Deviance: -473.5697
## AIC: -403.5697
## SBC: -281.7473
## ******************************************************************
getSmo(misto_normalp_temp_team)
## Linear mixed-effects model fit by maximum likelihood
## Data: Data
## Log-likelihood: 236.7848
## Fixed: fix.formula
## (Intercept)
## -7.596768e-18
##
## Random effects:
## Formula: ~1 | Numero_temporada
## (Intercept) Residual
## StdDev: 9.839206e-07 0.9999999
##
## Variance function:
## Structure: fixed weights
## Formula: ~W.var
## Number of Observations: 240
## Number of Groups: 15
#Resíduos
plot(misto_normalp_temp_team)
## ******************************************************************
## Summary of the Quantile Residuals
## mean = -2.035307e-18
## variance = 1.004184
## coef. of skewness = -0.4382711
## coef. of kurtosis = 3.51579
## Filliben correlation coefficient = 0.9918678
## ******************************************************************
shapiro.test(misto_normalp_temp_team$residuals) #p-value =
##
## Shapiro-Wilk normality test
##
## data: misto_normalp_temp_team$residuals
## W = 0.98337, p-value = 0.006562
#Independência
plot(misto_normalp_temp_team$residuals,
ylab = "Residuos",
xlab = "Index dos Imovéis",
main = "Suposição de independência",
pch = 19)
#Breusch_Pagan para homocedasticdade
bptest(misto_normalp_temp_team) #p-value =
##
## studentized Breusch-Pagan test
##
## data: misto_normalp_temp_team
## BP = 60.063, df = 34, p-value = 0.003813
##### Beta Team ####
misto_betap_ftp <- gamlss(formula = WINP ~ (re(random = ~1 | TEAM)) +
PlusMinus + PF + BLKA + FTP, family = BEZI, data = dados_regressaop)
## GAMLSS-RS iteration 1: Global Deviance = -170.0295
## GAMLSS-RS iteration 2: Global Deviance = -349.6334
## GAMLSS-RS iteration 3: Global Deviance = -350.6895
## GAMLSS-RS iteration 4: Global Deviance = -350.6399
## GAMLSS-RS iteration 5: Global Deviance = -350.6347
## GAMLSS-RS iteration 6: Global Deviance = -350.6342
misto_betap_ftp
##
## Family: c("BEZI", "Zero Inflated Beta")
## Fitting method: RS()
##
## Call: gamlss(formula = WINP ~ (re(random = ~1 | TEAM)) +
## PlusMinus + PF + BLKA + FTP, family = BEZI, data = dados_regressaop)
##
## Mu Coefficients:
## (Intercept) re(random = ~1 | TEAM) PlusMinus
## 0.215798 NA 0.103393
## PF BLKA FTP
## -0.025465 -0.028687 0.005414
## Sigma Coefficients:
## (Intercept)
## 3.661
## Nu Coefficients:
## (Intercept)
## -2.197
##
## Degrees of Freedom for the fit: 13.58 Residual Deg. of Freedom 226.4
## Global Deviance: -350.634
## AIC: -323.47
## SBC: -276.196
coef(misto_betap_ftp)
## (Intercept) re(random = ~1 | TEAM) PlusMinus
## 0.215797839 NA 0.103392931
## PF BLKA FTP
## -0.025464733 -0.028687176 0.005413543
summary(misto_betap_ftp) #AIC:
## ******************************************************************
## Family: c("BEZI", "Zero Inflated Beta")
##
## Call: gamlss(formula = WINP ~ (re(random = ~1 | TEAM)) +
## PlusMinus + PF + BLKA + FTP, family = BEZI, data = dados_regressaop)
##
## Fitting method: RS()
##
## ------------------------------------------------------------------
## Mu link function: logit
## Mu Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.215798 0.482300 0.447 0.6550
## PlusMinus 0.103393 0.004502 22.968 <2e-16 ***
## PF -0.025465 0.010889 -2.339 0.0202 *
## BLKA -0.028687 0.017780 -1.613 0.1080
## FTP 0.005414 0.004933 1.097 0.2737
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ------------------------------------------------------------------
## Sigma link function: log
## Sigma Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.66081 0.09514 38.48 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ------------------------------------------------------------------
## Nu link function: logit
## Nu Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -2.1972 0.2152 -10.21 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ------------------------------------------------------------------
## NOTE: Additive smoothing terms exist in the formulas:
## i) Std. Error for smoothers are for the linear effect only.
## ii) Std. Error for the linear terms maybe are not accurate.
## ------------------------------------------------------------------
## No. of observations in the fit: 240
## Degrees of Freedom for the fit: 13.58204
## Residual Deg. of Freedom: 226.418
## at cycle: 6
##
## Global Deviance: -350.6342
## AIC: -323.4701
## SBC: -276.1959
## ******************************************************************
getSmo(misto_betap_ftp)
## Linear mixed-effects model fit by maximum likelihood
## Data: Data
## Log-likelihood: -374.7157
## Fixed: fix.formula
## (Intercept)
## -0.0007885399
##
## Random effects:
## Formula: ~1 | TEAM
## (Intercept) Residual
## StdDev: 0.07091602 0.9627084
##
## Variance function:
## Structure: fixed weights
## Formula: ~W.var
## Number of Observations: 240
## Number of Groups: 33
#Resíduos
plot(misto_betap_ftp)
## ******************************************************************
## Summary of the Randomised Quantile Residuals
## mean = -0.1737727
## variance = 0.7273011
## coef. of skewness = -0.5256851
## coef. of kurtosis = 2.742173
## Filliben correlation coefficient = 0.9845265
## ******************************************************************
shapiro.test(misto_betap_ftp$residuals) #p-value =
##
## Shapiro-Wilk normality test
##
## data: misto_betap_ftp$residuals
## W = 0.96833, p-value = 3.504e-05
#Independência
plot(misto_betap_ftp$residuals,
ylab = "Residuos",
xlab = "Index dos Imovéis",
main = "Suposição de independência",
pch = 19)
#Breusch_Pagan para homocedasticdade
bptest(misto_betap_ftp) #p-value =
##
## studentized Breusch-Pagan test
##
## data: misto_betap_ftp
## BP = 19.939, df = 4, p-value = 0.0005133
##### Beta Temp ####
misto_betap_temp <- gamlss(formula = WINP ~ (re(random = ~1 | Numero_temporada)) +
PlusMinus + PF, family = BEZI, data = dados_regressaop)
## GAMLSS-RS iteration 1: Global Deviance = -161.4236
## GAMLSS-RS iteration 2: Global Deviance = -329.2529
## GAMLSS-RS iteration 3: Global Deviance = -330.669
## GAMLSS-RS iteration 4: Global Deviance = -330.6692
misto_betap_temp
##
## Family: c("BEZI", "Zero Inflated Beta")
## Fitting method: RS()
##
## Call: gamlss(formula = WINP ~ (re(random = ~1 | Numero_temporada)) +
## PlusMinus + PF, family = BEZI, data = dados_regressaop)
##
## Mu Coefficients:
## (Intercept) re(random = ~1 | Numero_temporada)
## 0.5145 NA
## PlusMinus PF
## 0.1063 -0.0265
## Sigma Coefficients:
## (Intercept)
## 3.571
## Nu Coefficients:
## (Intercept)
## -2.197
##
## Degrees of Freedom for the fit: 4 Residual Deg. of Freedom 236
## Global Deviance: -330.669
## AIC: -322.669
## SBC: -308.747
coef(misto_betap_temp)
## (Intercept) re(random = ~1 | Numero_temporada)
## 0.51448416 NA
## PlusMinus PF
## 0.10628690 -0.02649885
summary(misto_betap_temp) #AIC:
## ******************************************************************
## Family: c("BEZI", "Zero Inflated Beta")
##
## Call: gamlss(formula = WINP ~ (re(random = ~1 | Numero_temporada)) +
## PlusMinus + PF, family = BEZI, data = dados_regressaop)
##
## Fitting method: RS()
##
## ------------------------------------------------------------------
## Mu link function: logit
## Mu Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 0.514484 0.244311 2.106 0.0363 *
## PlusMinus 0.106287 0.004552 23.349 <2e-16 ***
## PF -0.026499 0.011248 -2.356 0.0193 *
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ------------------------------------------------------------------
## Sigma link function: log
## Sigma Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 3.57084 0.09504 37.57 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ------------------------------------------------------------------
## Nu link function: logit
## Nu Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) -2.1972 0.2152 -10.21 <2e-16 ***
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## ------------------------------------------------------------------
## NOTE: Additive smoothing terms exist in the formulas:
## i) Std. Error for smoothers are for the linear effect only.
## ii) Std. Error for the linear terms maybe are not accurate.
## ------------------------------------------------------------------
## No. of observations in the fit: 240
## Degrees of Freedom for the fit: 4
## Residual Deg. of Freedom: 236
## at cycle: 4
##
## Global Deviance: -330.6692
## AIC: -322.6692
## SBC: -308.7466
## ******************************************************************
getSmo(misto_betap_temp)
## Linear mixed-effects model fit by maximum likelihood
## Data: Data
## Log-likelihood: -376.032
## Fixed: fix.formula
## (Intercept)
## 4.471334e-12
##
## Random effects:
## Formula: ~1 | Numero_temporada
## (Intercept) Residual
## StdDev: 8.256872e-06 0.9485917
##
## Variance function:
## Structure: fixed weights
## Formula: ~W.var
## Number of Observations: 240
## Number of Groups: 15
#Resíduos
plot(misto_betap_temp)
## ******************************************************************
## Summary of the Randomised Quantile Residuals
## mean = -0.1821301
## variance = 0.7877213
## coef. of skewness = -0.907402
## coef. of kurtosis = 4.936006
## Filliben correlation coefficient = 0.9717185
## ******************************************************************
shapiro.test(misto_betap_temp$residuals) #p-value =
##
## Shapiro-Wilk normality test
##
## data: misto_betap_temp$residuals
## W = 0.94588, p-value = 8.96e-08
#Independência
plot(misto_betap_temp$residuals,
ylab = "Residuos",
xlab = "Index dos Imovéis",
main = "Suposição de independência",
pch = 19)
#Breusch_Pagan para homocedasticdade
bptest(misto_betap_temp) #p-value =
##
## studentized Breusch-Pagan test
##
## data: misto_betap_temp
## BP = 18.652, df = 2, p-value = 8.91e-05